For decades the neural processes underlying vestibular function were primarily studied in relationship to reflexes, but little has been done to understand its sensory functions, as they relate to self-motion perception and spatial orientation. Here we propose to remedy this void by applying signal detection and population decoding principles to vestibular afferent responses and how they might shape perception. We specifically target the otolith system and its role in heading (i.e., translation self-motion) and tilt perception. In aim 1, we will measure the 3D translation tuning of regular and irregular otolith afferents and use population decoding algorithms to compute neuronal and population direction discrimination thresholds. We hypothesize that (1) the vestibular periphery limits our ability to discriminate small differences in motion direction, with regular afferents having lower discrimination thresholds than irregular afferents; (2) optimal decoding of otolith afferent responses accounts for the observed dependence of perceptual heading discrimination thresholds on motion direction; (3) heading direction and tilt discrimination neuronal thresholds increase as a function of head orientation relative to gravity. In aim 2, we will compare heading direction discrimination thresholds of 2nd-order neurons in the vestibular nuclei with those of otolith afferents. We will test the hypothesis that VN discrimination thresholds are lower than those of otolith afferents. Such result would support sharpening of tuning and/or decrease in response variability by combining signals from both labyrinths and both sides of the striola. Finally, in aim 3, we will investigate whether noise in ololith afferents contributes to perception. While macaques perform a direction discrimination task, we will search for trial-to-trial correlations between otolith afferent activity and perception (choice probabilities, CPs). Such correlations, often found in cortical neurons, are thought to reflect top-down, feature-based attention influences. The present experiments seek to provide evidence for or against the bottom-up hypothesis, i.e., that sensory noise propagates through to the decision. Results from these studies are critical for modeling psychophysical data and understanding how the vestibular system contributes to perception. In addition, outcomes will contribute important knowledge to our current understanding about neuronal variability and how it shapes perception.